Used Car Price Predictor

A Random Forest model trained on ~17k used-car listings.

This model was trained on a public scraped used-car listings dataset for a data structures course project. On held-out test data it reaches R² ≈ 0.71 and a median error of ~14% - treat this as a rough estimate, not an appraisal. Accuracy is lowest for rare/luxury cars (under-represented in the training data) and can be thrown off by unusual combinations of inputs. See the Model info tab for the full breakdown.